|Home | About | Journals | Submit | Contact Us | Français|
The unlimited differentiation and proliferation capacity of embryonic stem cells represents a great resource for regenerative medicine. Here, we describe a method for differentiating, isolating, and expanding endothelial cells (ECs) from mouse embryonic stem cells (mESCs). First, mESCs are expanded on a mouse embryonic fibroblast (mEF) feeder layer and partially differentiated into embryoid bodies (EBs) by growing the cells in an ultra-low attachment plate for up to 5 days. The EBs are then differentiated along the endothelial lineage using endothelial growth medium supplemented with 40 ng/mL vascular endothelial growth factor (VEGF). The differentiated endothelial population expresses both Fetal Liver Kinase 1 (Flk-1) and VE-Cadherin on the cell surface which can be further purified using a fluorescence-activated cell sorting (FACS) system and subsequently expanded on 0.1 % gelatin-coated plates. The differentiated cells can be analyzed by real-time PCR and flow cytometry to confirm enrichment of EC-specific genes and proteins.
Embryonic stem cells (ESCs), which are derived from the inner cell mass of blastocysts, are capable of self-renewal and differentiation along different cell lineages, a property known as pluripotence (1). Because ESC differentiation can be directed into particular lineages with specialized functional properties for tissue repair and replacement, they are considered to be an excellent resource for regenerative medicine (2). Three major obstacles associated with using ESCs for regenerative medicine are: (1) precise and controlled differentiation of ESCs toward a well-defined lineage; (2) isolation of homogeneous populations of stably differentiated cells that are fully functional; and (3) retaining the expansion potential.
A common approach is to pre-differentiate ESCs into three-dimensional cell aggregates known as embryoid bodies (EBs). These bodies contain cells from all three germ layers, mesoderm, ectoderm, and endoderm, and can self-renew and differentiate into different cell types. Embryoid bodies can be produced in three main ways: (1) hanging drop; (2) methylcellulose hydrogel; and (3) suspension culture. The most convenient and efficient of these is suspension culture of ESCs in non-adherent plates (3, 4). This approach produces large amounts of EBs in a short time, which is advantageous for high-throughput drug screening and tissue engineering. However, an important consideration is embryoid body size, which may affect the outcome and properties of differentiated cells. For example, uncontrolled overgrowth of EBs may result in cavity formation due to apoptosis, after which EBs eventually become cystic and contain fluid.
Methods for developing differentiated endothelial cells from EBs are well established and the corresponding gene expression patterns are well characterized (4– 6). Embryonic stem cells have been directly differentiated toward specific lineages without first making EBs by using conditioned medium for endothelial differentiation of ESC in collagen IV plates (7, 8).
Large-scale changes in gene expression accompany the initial differentiation of ESCs into EBs, and subsequent large-scale changes in gene expression are linked to lineage-specific differentiation of EBs along mesenchymal, epithelial, neural, or hematopoietic lineages (9, 10, 11). Embryoid bodies can be directed to give rise to hemangioblasts, which subsequently undergo further differentiation into either hematopoietic or endothelial cells. Hemangioblasts have been widely used to study the expression of transcription factors that control EC lineage and recapitulate many aspects of vascular development in vivo (12). The expression of vascular endothelial growth factor receptor-2 (VEGF-R2), also known as Flk-1, in mice, is a mesodermal indicator and the earliest functional marker for hemangioblasts (13).
After EBs are formed and exposed to endothelial differentiation medium containing VEGF, a heterogeneous cell population, including mesenchymal, hematopoietic, and epithelial cells, emerges. Only a subset of these cells will differentiate toward endothelial cells (less than 2 % with our current method). Therefore, once differentiation has been induced the endothelial cell population must be isolated and purified for further expansion and analysis.
Herein, we describe our methods for endothelial differentiation of mouse ESCs (mESCs) and FACS-based isolation and expansion of the resulting cell populations that express the endothelial-specific markers Flk-1 and VE-Cadherin on their surface.
One method to expand mESCs is to culture them on mEF feeder layers that have been inactivated with Mitomycin C or via irradiation and with the differentiation inhibitory factor, Leukemia inhibitory factor (LIF) (14). When mEF cells are inactivated, the mESCs can attach, form clumps, and use the growth factors produced from feeder layers, while mEF cell proliferation is inhibited. Once expanded LIF and stromal contact are both withdrawn, mESCs can be grown in ultra-low attachment plates and partially differentiated to form EBs containing all germ layers. The EBs are subsequently differentiated toward the endothelial cell lineage by growing in gelatinized plates and exposing the cells to differentiation medium containing VEGF. The differentiated ECs can be selected via cell sorting with endothelial-specific cell surface antigens.
This method is for splitting the mESCs once they have been cultured on the mEF feeder layer. The mESCs will begin to form colonies on the mEF feeder layer that can eventually merge. Because mESCs only maintain their undifferentiated state when colonies are not merged, cells must be passaged before colonies come in contact with each other.
After treating the mESCs and mEFs cells with 0.1 % collagenase IV (see Subheading 3.2.2), mESCs will primarily detach; however, some mEF cells may also detach from the plate. As only mESCs will be used for EB formation, mESCs can be separated from detached mEF cells by allowing the mEF cells to differentially adhere onto new 10 cm culture dishes.
Subculture all cells (mESC + mEF) onto a new 10 cm plate using mESC medium and incubate at 37 °C and 5 % CO2 for 0.5–1 h. The mEF cells will attach during this time and mESCs will stay in suspension. After 0.5–1 h, transfer the non-adherent mESCs into a 15 mL falcon tube and triturate the cells using a 5 mL pipette to break down the colonies. These cells will be used to prepare EBs.
The size of EBs is an important factor that can affect properties of differentiated cells. The mean size of EBs after 5–6 days of culture in ultra-low attachment plates described above should be around 100–250 µm (approximately to 10–25 cells wide). Smaller EBs might not contain all three germ layers, whereas overgrown EBs will start undergoing apoptosis (Fig. 1).
Differentiation starts when EBs form (while in suspension culture, see Subheading 3.4) and will continue after the EBs adhere to the 0.1 % gelatinized plates. The cells adhering and growing out from the attached EBs have an extended morphology at early time points (day 1–2) in gelatinized plate (Fig. 2A) and become more rounded at later time points (day 7) (Fig. 2D).
In the early stages of differentiation, the culture conditions to promote endothelial differentiation described above give rise to a heterogeneous population with less than 2 % of the cells being pure endothelial lineage. Thus, to study the characteristics and fate of purified endothelial progenitors, this small subpopulation of cells must be isolated from the mixed cultures. FACS is a specialized type of flow cytometry for separating heterogeneous mixtures of biological cells into multiple fractions, one cell at a time, based on light scattering and fluorescent characteristics of each cell. Lasers are used to excite intrinsic or extrinsic fluorescence of cells and the fluorescence intensity is measured from cells or particles through sensitive photomultiplier tubes (15, 16).
The combination of flow cytometry and single-cell sorting is a powerful way to identify and isolate cells with particular characteristics, for instance, based upon markers expressed on the cell surface during differentiation. Control setup and data validation can sometimes be complex, especially when cells have a high fluorescence background or are transfected with a fluorescence reporter before being labeled. Nevertheless, compared to other techniques, FACS facilitates rapid data acquisition, specific multiparameter analysis, and functional separation with high accuracy (17, 18).
During the cell sorting process, a tunable transducer permits the fluid sheath to be broken into individual droplets such that each droplet encapsulates single cells. An electric circuitry places an electrical charge on the fluid stream and the individual droplets. The point at which a cell passes through the laser focus and enters into a droplet corresponds to a specific delay. Because the droplets carry a charge on their surface, a de flecting plate can redirect these charged droplets to collection tubes. Sorting criteria, region designation, multiparameter acquisition, and/or analysis are defined by a software system that includes display platforms.
An alternative to FACS system for cell sorting is magnetic cell separation. This technique is based on magnetic labeling of cells with very small microbeads that do not alter cell structure and function of cells. Separation of labeled cells takes place within a column that provides a magnetic field for cell sorting.
As mentioned above, conditions that support endothelial differentiation of mESCs yield a heterogeneous population of cells with less than 2 % pure endothelial cells. Although hemangioblasts, which express Flk-1, can differentiate into both hematopoietic and endothelial cells, endothelial progenitor cells also express VE-Cadherin. Thus, to distinguish endothelial precursor cells from hemangioblasts, mesenchymal, epithelial, and other mixed cell types within the differentiating cultures, selecting for cells expressing both Flk-1 and VE-Cadherin is advisable (see Note 9). By labeling putative ECs with two fluorescent-tagged antibodies against Flk-1 and VE-Cadherin, ECs can be selected via FACS. We have used mESCs that were transfected with green fluorescent protein (GFP), which required initial compensation (18–20 %), to de fine negative populations before sorting by Flk-1 and VE-Cadherin surface markers.
Samples can then be analyzed on a Becton Dickinson FACS Vantage TM/DIVA, with a nozzle size of 80 µm, in which forward light scatter (FSC) is collected through a neutral density filter in the forward light scatter path, and side scatter (SSC) is collected through a neutral density filter at a 90° angle. The 488 nm lasers excite fluorescein isothiocyanate (FITC)/GFP and PE, while the 633 nm laser excites APC. Fluorescence emissions can be collected through the FITC (533/30 BP), PE (585/42 BP), and APC (660/20 BP) filters in fluorescence channels FL1, FL2, and FL4, respectively. Below are the steps for sample preparation for FACS sorting and expansion of sorted cells:
All the steps related to endothelial differentiation of mouse embryonic stem cells have been summarized in Fig. 4.
This work is supported by grants from NIH/NCI TMEN grant (U54CA126552.) to Nancy Boudreau and Mina J Bissell and U.S. Department of Energy, Office of Biological and Environmental Research (DE-AC02-05CH1123), a Distinguished Fellow Award and Low Dose Radiation Program (03-76SF00098) to Mina J. Bissell. Mandana Veiseh was supported by a postdoctoral fellowship from the NCI of the NIH (F32 CA132491A). We thank Pamela Derish in the Department of Surgery at UCSF for editorial review of the manuscript.
1Mitomycin C solution is light sensitive. Store the mixture at 4 °C in a dark container for up to 6 weeks or at −20 °C for up to 4 months.
2Freezing medium is light sensitive. Turn off the tissue culture hood lights during sample preparation. Aliquot the freezing medium into smaller batches (3–5 mL) and keep at 4 °C in a foil-covered container for up to 2 weeks, or store at −20 °C for up to 4 months.
3Differentiation medium is light sensitive. Cover the medium container with aluminum foil. Also prepare fresh medium in smaller batches for each experiment and store at 4 °C for up to 4 weeks.
4After dissolving the Collagenase IV in Knockout D-MEM medium, filter the solution using a 0.2 µm filter, then aliquot the solution into smaller batches (3–5 mL), and store at −20 °C. Samples can be kept up to 4 months.
5To store poly-HEMA coated plates, wrap them in parafilm after exposing them to UV light and keep them at room temperature. Wash the plates three times with sterile D-PBS in a tissue culture hood before use.
6Cultured, mitotically-inactivated mEF cells can be kept for up to 2 weeks in mEF medium while changing the medium every 2–3 days.
7Primary mouse embryo fibroblasts, Hygro-resistant strain C57BL/6 (Millipore, Billerica, MA) also can be used as an alternative to mEF cells. These primary cells have been treated with Mitomycin-C and can be directly plated onto gelatin-coated culture vessels. 2–3 × 106 cells will produce the appropriate density feeder layer on a 10 cm plate.
8It is important to monitor the cells and change their medium every day. mESCs can easily become contaminated.
9One potential way of improving the percentage of ECs differentiated from ESCs is to first sort the Flk-1+ cells and then continue differentiation of this subpopulation on collagen type IV coated plate prior to sorting for endothelial cells. The percentage of double-positive cells for VE-Cadherin and Flk-1 can be increased up to fivefold.
10Perform the identical steps for staining and sample preparation on all control samples.
11Control samples are used for de fining the background/negative signal, compensation values, and multiparameter sort criteria. Sort region is defined on the double-positive quadrant of a double-stained sample.
12The EC differentiation medium can be switched to EGM-2 medium after 1 week.